| | ENG

For Investigators Only

Recovery from Spinal Cord Injury

Wise Young, Ph.D. M.D.

Many people and animals recover from devastating injuries that must have destroyed as much as 90% of the spinal tracts at the injury site. I have long felt that if we understood and can implement the mechanisms by which people and animals recover from spinal cord injury, we would have a cure. How and why do people and animals recover from spinal cord injury?

Recovery is the Rule and Not the Exception after Spinal Cord Injury

The importance of the spinal cord to survival is clear from the evolution of the complex, beautiful, and flexible vertebral column that protects the spinal cord against trauma. As many as a million people suffer trauma to the spinal column every year in the United States but only 1% (10,000 people) have serious neurological deficits. We have many names for spinal injuries that produce only transient neurological deficits: whiplash, stingers, burners, etc. Although most clinicians assume that the spinal cord has not been damaged in such accidents, it is possible that many people may have suffered some spinal cord injury but simply recovered. Over 60% of people with severe spinal cord injuries have some residual motor or sensory function below the injury site shortly after injury. Most of these patients recover locomotor, sensory, bladder, and other functions over a period of many months.

My first encounter with a person who recovered from spinal cord injury was in 1981. Carey Erickson was a well-known choreographer in New York. He injured his cervical spinal cord at C4-5 and had no function below his injury site except for a patch of sensation on his left leg when he was admitted. I did weekly somatosensory evoked potentials on him for 72 days and saw his response improve from almost nothing to virtually normal in amplitude and latency. Over several months, he recovered his legs first and then his arms. At 2 years after injury, he walked into my office and, to most casual observers, appeared to have recovered completely. He, however, told me that his strength, coordination, and endurance were only 50% of what they were before his injury.

Another example is a young woman who fell off a horse and had a C5 cervical spinal cord injury. Although her father noted that she had some sensation in the legs when she was loaded into the ambulance, she had no movement or feeling in all four limbs for weeks. Over a period of 3 months, however, she recovered substantially. When I saw her 3 months after injury, she walked into the room with a cane. Later, she accompanied me on a visit to a lecture hall and we walked up a steep hill together. Although she still had to be careful when she walked, her endurance and strength was good. She has recovered virtually completely.

Many athletes have made remarkable recoveries after spinal cord injury. For example, Dennis Byrd of the New York Jets walked out of the hospital. He, like Carey Erickson, had only a patch of sensation on one leg at the time of admission to hospital. Reggie Brown of the Detroit Lions is another football player who walked out of the hospital after a cervical spinal cord injury that had paralyzed his arms and legs. I met him at a function several years later and, if I did not know his story, I would not have suspected that he had spinal cord injury. More recently, Adam Taliaferro, a freshman cornerback for Penn State, had a cervical spinal cord injury that paralyzed his arms and legs for several weeks. He is walking.

Dozens of people have come to me and told me of their recovery from quadriplegia to normality. Most had some residual motor or sensory function during the first 24 hours after injury. Almost all received methylprednisolone shortly after injury. The recovery took place over weeks, months, or even years after injury. There have not been any rigorous studies of the incidence of dramatic recoveries after spinal cord injury. In 1992, I surveyed some 400 patients whom we cared for at Bellevue Hospital in the 1980's. During those years, we were already treating people with methylprednisolone. Some 17% of the patients walked out of the hospital. Admittedly, most were "incomplete" injuries. But, some were "complete" spinal cord injuries.

In the NASCIS 2 trial carried out in the late 1980's, over 60% of the patients had complete loss of motor function and untreated patients recovered on average 8% of the motor score that they had lost, compared to 21% in patients that received methylprednisolone with 8 hours. So-called "incomplete" untreated patients recovered 59% of what they had lost by 6 months, compared to 75% in methylprednisolone-treated patients. In the recent study by Geisler, et al., (2001), 29% of nearly 800 patients showed "marked recovery" (i.e. a 2-category improvement on the Benzel score). Approximately 17% of the ASIA A ("complete" spinal cord injuries) patients had "marked recovery" by 52 weeks after injury. By comparison, about 40% of ASIA B ("sensory preservation only") and about 75% of ASIA C ("some motor and sensory preservation) patients had "marked recovery".

Thus, it is fair to say that recovery is the rule and not the exception in spinal cord injury. The observation that 17% of patients with so-called "complete" spinal cord injury had "marked recovery" of 2 categories on the Benzel scale is impressive. Close to 40% of patients with only minor sensory preservation had marked recovery. Finally, over 75% of patients with some motor and sensory preservation recovered walking. As pointed out above, many people recover almost completely from apparently severe spinal cord injuries that left them quadriplegic for weeks or even months. These data contrast sharply with the deep pessimism that the general public and many clinicians view the likelihood of recovery from spinal cord injury.

Recovery after Hemisection

The ability of the spinal cord to recover from injury is truly amazing if you consider that animals (and humans) will usually recover almost completely from a lateral hemisection, i.e. cutting a half of the spinal cord on one side. A hemisection eliminates half of the connections to and from the brain. In humans, this type of injury produces a Brown-Sequard syndrome, named after a British neurologist who described the syndrome in patients who had been punished by the Mafia with a stiletto inserted into the spinal cord to cut half the spinal cord on one side. There is an initial period of paralysis and sensory loss on the side of the cut, with loss of pain and temperature sensation on the other side. Over a period of several months, most people will recover ability to walk with both legs although sensory recovery may take many years and, in some patients, do not recover.

In animals, a hemisection likewise causes an initial period of neurological loss but most rats will recover ability to walk with both hindlimbs 2 weeks after a lateral hemisection. In a relatively little-known paper published, Kato, et al. (1985) described the remarkable recovery of cats from bilateral hemisections. They hemisected the spinal cord at T12, waited 0-126 days, and then hemisected the other side of the spinal cord at T7. Incredibly, when the hemisections were separated by 10 or more days, the cats recovered walking even though all the tracts between the brain and spinal cord have been severed. If the two hemisections were carried out in one surgery, the cats remained permanently paralyzed. If the hemisections were separated by 1-7 days, the cats required an average of 43 days to recover standing. If the second hemisection occurred 10-126 days after the first, the cats required an average of 15 days to recover standing. Over a period of several months, most of the latter recover bipedal walking. Bilateral hemisections should eliminate all long tracts between the brain and the lower spinal cord.

What are some of the mechanisms of recovery that could be playing a role in recovery? Animal studies suggest at least two mechanisms of recovery:

  • Sprouting for preserved contralateral spinal tracts. In 1994, Saruhashi, et al. examined the locomotor recovery in rats after hemisection, looking specifically at serotonergic fibers that crossed the midline from the intact side. All the rats recovered bipedal locomotion within 2-4 weeks and the extent of recovery correlated linearly with growth of serotonergic fibers from the intact side to the lesioned side. Similar correlations of forelimb recovery with sprouting of the contralateral corticospinal tract occur after a unilateral corticospinal tract lesion (Z'Graggen, et al., 2000) or sprouting of the ventral corticospinal tract after a bilateral corticospinal tract lesion (Weidner, et al. 2001).

  • Multisynaptic pathways. In the 1980's, Alstermark examined the recovery of forepaw function in cats after cutting various spinal tracts (summarized in Peterson, et al. 1997). They found that the cats would recover almost completely from lesions of every individual tract, as long as they left the propriospinal tract (which mediates multisynaptic connections within the spinal cord) intact. However, when they combined lesions of the propriospinal tract with other tracts, they found deficits.

Probably a combination of both mechanisms play a role in recovery of locomotion after bilateral hemisection. The remarkable results of Kato, et al. (1985) could be explained by sprouting of axons across the midline which then activate locomotor centers in the lower spinal cord through multisynaptic propriospinal pathways. In 1990, Kato, et al. not only hemisected the spinal cord at L2-3 but also did a mid-line myelotomy (cutting the spinal cord down the midline) which isolated the lower left lumbar cord. After the lesion, the cats initially recovered ability to stand on three legs. However, two days later, the leg innervated by the isolated lower left lumbar cord recovered walking capabilities. Although the phase relationships of the walking were more variable than normal, the two legs stepped alternatively. Kato did not report histological examinations of the spinal cords but one possibility is sprouting across the midline.

Recovery after Contusion

The vast majority of human spinal cord injuries involve contusion or compression rather than physical transection. Contusion injury produces a stereotyped central hemorrhagic necrosis that usually leaves a thin rim of white matter at the injury site. In 1986, Blight & DeCrescito did a detailed and quantitative analysis of the surviving axons that are necessary and sufficient to support locomotor recovery in cats after a severe contusion injury. A 20 gram weight dropped 20 cm onto the thoracic spinal cord of a cat typically left a thin rim of less than 0.3 mm white matter. Quantitative counts of the axons revealed that cats with as little as 10% of their axons were able to recover independent locomotion. Since central hemorrhagic necrosis eliminated all deeper tracts, including the propriospinal tracts, the mechanism of recovery likely involved axons close to the pial surface.

Basso, et al. (1996) carried out a similar morphological correlation with locomotor recovery in rats after contusion injuries. Using the newly developed BBB locomotor scale, they showed that the degree of spinal cord white matter sparing correlated linearly with the BBB scores. Of interest, however, was the fact that when BBB scores were plotted against spared white matter (X-axis), the line intersected the Y-axis at a BBB score of about 8. A BBB score of 10 signifies a rat that is able to stand and take steps. Rats with only 10% spared white matter were able to walk. This is consistent with the experience of many other investigators.

In 1997, Beattie, et al. examined the spinal cords of several hundred rats that had received contusion injuries. Over 70% of the rats had cysts at the contusion site that was filled with a cellular matrix. Thousand of axons were growing on the cellular matrix. Although the origin and destination of the axons were not known, they were clearly regenerating axons. More recently, Hill, et al. (2001) showed that some of the fibers came from the corticospinal tract and reticulospinal tract. The extent to which these regenerating axons contributed to locomotor recovery is not clear. However, in earlier studies, Beattie, et al. transected the spinal cord and found that rats recovered BBB scores of 3-4. These scores imply that the rats were able to move 1 or 2 joints.

Could spontaneous regeneration be occurring in the rat spinal cord? Clearly, some regeneration did occur even though the functional significance of the regeneration has not yet been proven. Substantial and convincing data support an important role of axonal sprouting unilateral and across the midline in both forelimb and hindlimb recovery after hemisections. The spinal cord is clearly able to activate locomotor centers with less than 10% of the white matter. The remarkable capability of the spinal cord to recover locomotion after a bilateral hemisection and even after a hemisection plus midline myelotomy provides strong evidence of the robust capabilities of sprouting. If such amazing recovery can occur through sprouting under such adverse conditions, why can't we consider regeneration as a possibility?

What is the evidence that the spinal cord cannot regenerate? This dogma arose many decades ago when scientists were transecting the spinal cord. When the spinal cord is transected, the tension in the spinal cord usually forces the cut ends of the spinal cords apart. Few researchers took the trouble or had the ability to re-oppose the cut ends of the spinal cord. It is not so surprising that none of these studies revealed regeneration. After all, if there is one obstacle that axons are unlikely to surmount, it is the presence of a gap of several mm. The contusion model of spinal cord injury leaves continuous tissue. Studies of the contusion model indicate that axons are growing into and probably across the injury site. Thus, the possibility of spontaneous regeneration contributing to functional recovery after spinal cord injury should be seriously considered.

Finally, many people with spinal cord injury recover sensation and even motor function years after spinal cord injury. For example, Christopher Reeve suffered a C1/2 injury. He was examined by dozens of doctors and was as a "complete" spinal cord injury as ever documented. Yet, at about two years after injury, he recovered light touch sensation in his arms and hands, extending all the way to the lowest sacral levels. He is now able to go off the ventilator for several hours at a time, can shrug his shoulders (indicative of some C4 function), and apparently has shown some hand motion. The last appeared more than five years after injury. This kind of recovery occurring steadily and starting with more proximal levels and extending distally over time is strongly suggestive of regeneration. The timing of the recovery is consistent with the slow growth of axons in the spinal cord. We should keep our minds open to the possibility of spontaneous regeneration even in humans.


  • Kato M, Murakami S, Hirayama H and Hikino K (1985). Recovery of postural control following chronic bilateral hemisections at different spinal cord levels in adult cats. Exp Neurol. 90 (2): 350-64.
  • Saruhashi Y, Young W and Perkins R (1996). The recovery of 5-HT immunoreactivity in lumbosacral spinal cord and locomotor function after thoracic hemisection. Exp Neurol. 139 (2): 203-13.
  • Z'Graggen WJ, Fouad K, Raineteau O, Metz GA, Schwab ME and Kartje GL (2000). Compensatory sprouting and impulse rerouting after unilateral pyramidal tract lesion in neonatal rats. J Neurosci. 20 (17): 6561-9.
  • Weidner N, Ner A, Salimi N and Tuszynski MH (2001). Spontaneous corticospinal axonal plasticity and functional recovery after adult central nervous system injury. Proc Natl Acad Sci U S A. 98 (6): 3513-8.
  • Pettersson LG, Lundberg A, Alstermark B, Isa T and Tantisira B (1997). Effect of spinal cord lesions on forelimb target-reaching and on visually guided switching of target-reaching in the cat. Neurosci Res. 29 (3): 241-56.
  • Kato M (1990). Chronically isolated lumbar half spinal cord generates locomotor activities in the ipsilateral hindlimb of the cat. Neurosci Res. 9 (1): 22-34.
  • Blight AR and Decrescito V (1986). Morphometric analysis of experimental spinal cord injury in the cat: the relation of injury intensity to survival of myelinated axons. Neuroscience. 19 (1): 321-41.
  • Basso DM, Beattie MS and Bresnahan JC (1996). Graded histological and locomotor outcomes after spinal cord contusion using the NYU weight-drop device versus transection. Exp Neurol. 139 (2): 244-56.
  • Beattie MS, Bresnahan JC, Komon J, Tovar CA, Van Meter M, Anderson DK, Faden AI, Hsu CY, Noble LJ, Salzman S and Young W (1997). Endogenous repair after spinal cord contusion injuries in the rat. Exp Neurol. 148 (2): 453-63.
  • Hill CE, Beattie MS and Bresnahan JC (2001). Degeneration and sprouting of identified descending supraspinal axons after contusive spinal cord injury in the rat. Exp Neurol. 171 (1): 153-69.

To read more about spinal cord injury:

What is the Spinal Cord

What is Spinal Cord Injury

Spinal Cord Injury Levels and Classification

Acute Spinal Cord Injury

Chronic Problems of Spinal Cord Injury

Recovery and Treatment

Recovery from Spinal Cord Injury

Spinal Cord Injury and Family

Ten Frequently Asked Questions Concerning Cure of Spinal Cord Injury

Copyright © 2007. CHINASCINET
About Us | SCI Info | Studies | Events | ISCITT | Contact